P. Thambidurai, K.S. Trivedi, Transient Overloads in Fault-Tolerant Real-Time Systems, Real Time Systems Symposium, Santa Monica, Californie, 1989, pp 126-133.  Home/Search   Document Not in Database   Summary   Related Articles   Check

....failure probability is greater. But, active redundancy provides a fast recovery failure, because it asks no state s restoration (all the time, copies do the same process and are in the same state) This method is currently used for reliable real time systems where the recovery time is bounded [10]. 3.2. Passive redundancy A redundant element can only start running when the primary element is failing. All the time, only one element is active. I result data [ element 1 I I failure : etectim element 2 [ figure 8: passive redundancy When a copy replaces the faulty primary element, a ....

P. Thambidurai, K.S. Trivedi, Transient Overloads in Fault-Tolerant Real-Time Systems, Real Time Systems Symposium, Santa Monica, Californie, 1989, pp 126-133.

....than choosing to maintain optimal service to some views, thereby e ectively relieving the starvation. The EDF (Earliest Deadline First) scheduling algorithm [1] is applied in a wide range of real time systems and has been shown to be optimal with respect to di erent metrics in the literature [2, 8, 11]. However, in overload conditions, it performs very poorly because EDF gives the highest priority to those tasks that are close to missing their deadlines. Therefore, a typical phenomenon that may happen with EDF when the system is overloaded is the domino e ect , since the rst task that misses ....

....the overload interval. To the best of our knowledge, previous work [2, 14, 3, 9, 15] focused on the case when EPU is the measure of scheduling performance. Some algorithms have also been proposed to deal with overloads that improve the performance of EDF that is optimal when system is underloaded [2, 8, 11]. Baruah et al. 13] have shown that there exists an upper bound on the performance of any on line (preemptive) algorithm working in overload conditions. Traditionally, the goodness of an online algorithm working in overload conditions is measured with respect to a clairvoyant scheduler (one ....

P. Thambidurai, K. S. Trivedi. Transient overloads in fault-tolerant real-time systems. In Proceedings of Real-Time Systems Symposium, December 1989.

....to identify the faulty module. Once identified, the faulty module can be replaced with a backup module. In DRAFT, the outputs from these fault tolerance methods are produced by hardware instead of software. Because there are continuing constraints arising from both performance and fault tolerance [6], the treatment of these outputs could be done in hardware or software. 3 Selecting Applying FPGAs A FPGA generally consists of a two dimensional array of logic blocks (or cells) that can be connected through interconnection resources. Cells in Atmel s FPGAs [7] can perform a combinational ....

Thambidurai, P., and Trivedi, K., "Transient Overloads in Fault-Tolerant Real-Time Systems," Proceedings of the Real-Time Systems Symposium, IEEE Computer Society Press, Dec. 1989, pp. 126-133.

....a task is a complex and sometimes inaccurate procedure. This section examines how global scheduling performs when the worst case computation time is inaccurate (i.e, can be over or under estimated) and some deadlines can be missed. This should typically increases the utilization of the processors [10, 11]. 4.1 Admission control Although some tasks may sometimes miss a deadline in a soft real time system, an admission control is still useful. This admission control should reject tasks that are likely to miss their deadlines. The experiments from Section 3 have shown that the processor utilization ....

P. Thambidurai and K.S. Trivedi. Transient overloads in fault-tolerant real-time systems. In Real-Time Systems Symposium, pages 126--133, Dec 1989.

....in practical systems, since in real world applications intermittent overloads may occur due to exceptional situations, such as modifications in the environment, arrival of a burst of tasks, or cascades of system failures. A number of heuristic algorithms has been proposed to deal with overloads [2, 3, 4, 6, 7, 10, 11, 13, 15]. They all improve the performance of EDF, however, very few simulation studies have been done to evaluate the importance of the scheduling policy with respect to the guarantee mechanism used to deal with the overload. Baruah et al. 1] have shown that there exists an upper bound on the ....

P. Thambidurai and K. S. Trivedi, "Transient Overloads in Fault-Tolerant Real-Time Systems," Proceedings of Real-Time Systems Symposium, December 1989.

.... A heuristic was proposed in the Spring project, where deadline and cost driven algorithms are combined together with rules to dynamically revise values and deadlines in accordance with the precedence relations [6] A number of heuristic algorithms have also been proposed to deal with overloads [30] [13] which improve the performance of EDF. Baruah, et al. 3] have shown that there exists an upper bound on the performance of any on line (preemptive) algorithm working in overload conditions. The goodness of an online algorithm is measured with respect to a clairvoyant scheduler (one that ....

P. Thambidurai and K. S. Trivedi, "Transient Overloads in Fault-Tolerant Real-Time Systems," Proceedings of Real-Time Systems Symposium, December 1989.

....for non preemptive tasks, rather than the preemptive model used here, but did not address overload. Other general work on overload in real time systems has also been done. For example, Sha [Sha 86] shows that the rate monotonic algorithm has poor properties in overload. Thambidurai and Trivedi [Tha 89] study transient overloads in fault tolerant real time systems, building and analyzing a stochastic model for such a system. However, they provide no details on the scheduling algorithm itself. Schwan and Zhou [Sch 92] do on line guarantees based on keeping a slot list and searching for free time ....

P. Thambidurai, and K. S. Trivedi, "Transient Overloads in Fault-Tolerant RealTime Systems," Proceedings of Real-Time Systems Symposium, December 1989.

No context found.

Thambidurai, P., Trivedi, K. S., Transient Overloads in Fault-Tolerant Real-Time Systems, IEEE Real-Time Systems Symposium, 1989. 24

No context found.

Thambidurai, Phillip and Kishor S. Trevidi. Transient Overloads in FaultTolerant Real-Time Systems. Proceedings of the IEEE Real-Time Systems Symposium (December 1989) vol. 10, pp. 126-133.

Online articles have much greater impact   More about CiteSeer.IST   Add search form to your site   Submit documents   Feedback  

CiteSeer.IST - Copyright Penn State and NEC